Coronary atherosclerosis is the major cause of death in postmenopausal women in the USA. While coronary artery bypass surgery decreases symptomatic and clinial evidence of cardioischemia, it does not alter the underlying atherosclerotic process. Patients may develop recurrent symptoms due to saphenous vein graft occlusion, graft atherosclerosis, or progression of underlying coronary artery disease. Interventions that reduce the rate of coronary atherosclerosis progression would significantly benefit women following bypass surgery and possibly benefit all women with atherosclerotic disease. Some observational studies suggest that postmenopausal estrogen replacement therapy (HRT) reduces cardiac morbidity by up to 50%. However, the HERS Study published in 1998, showed no benefit in the clinical outcomes of sudden cardiac death or myocardial infarction in postmenopausal women with known coronary disease randomized to combined conjugated equine estrogen and medroxyprogesterone or placebo over 4 years of follow-up. This study also suggested an increase in these endpoints in the first years of treatment with the active drugs and a subsequent decrease in event rates compared to the placebo treatment with the active drugs and a subsequent decrease in event rates compared to the placebo arm. The efficacy of HRT to delay the development of saphenous graft atherosclerosis is unknown. This randomized, double-masked, placebo-controlled trial tests the hypothesis that HRT started shortly after coronary bypass surgery will delay the development of graft atherosclerosis and reduce the occurrence of graft occlusion. Women are randomized to placebo or HRT with 17 beta-estradiol plus medroxyprogesterone acetate (or 17 beta-estradiol alone if post hysterectomy) within 6 months of surgery. The development of vein graft atherosclerosis will be measured using quantitative coronary angiographic and intravascular ultrasound assessment of disease severity and extent. Studies will be performed 6 months and 3.5 years after randomization. We shall determine the influence of HRT on the primary outcome variables of the change in severity and extent of atherosclerosis in vein grafts over the 3.5 years of therapy. We postulate that the pathophysiologic mechanisms of platelet activation, fibrinogen binding to platelets, vascular reactivity, coagulation and fibrinolytic factors and lipoprotein composition predict the occurrence of graft occulsion and graft atherosclerosis. The effect of HRT on these factors will be masured. The proposal also tests the hypothesis that HRT exerts its beneficial effects on these risk factors in addition to more traditional risk factors including lipids and lipoprotein profile. The influence of these risk factors and the effect of HRT on the frequency of early graft closure (identified on a 6 month coronary angiogram) will be assessed. The study is a multicenter trial with the Data Coordinating Center and principal Investigator located at Johns Hopkins. Women are recruited from the University of Maryland Hospital, Johns Hopkins Hospital and followed at Johns Hopkins. The other sites include: Wake Forest University School of Medicine, Hartford Hospital, Christiana Hospital, Cleveland Clinic, University Hospitals of Cleveland, Montreal Heart Institute, Ottawa Heart Insititute, Washington Hospital Center, Miriam Hospital in Rhode Island, University of Connecticut, and the Indiana Heart Institute. Additional sites have been added because of the difficulty in recruitment experienced at every site.